In EPC (Evolved Packet Core), bearer management is performed based on an ‘Always On’ concept in such a manner that a PGW (PDN (Packet Data Network) Gateway), initially selected when a UE (User Equipment or ‘terminal’) has attached the EPC, is fixedly used as anchor until the UE detaches. By this manner of operation, it is possible for a service network (a packet data network) to provide services premised on permanent connection. It is because the IP information driven out by the PGW is unchanged even if the UE moves repeatedly within the EPC.
When a UE moves within the EPC, an SGW (Serving Gateway) is re-selected in accompaniment with the movement of the UE. Each time the SGW is re-selected, a bearer between the SGW and the PGW is updated by disconnection and re-establishment to ensure connectivity from the UE to the PGW.
In general, in selecting a PGW, such a PGW is selected which is close to the SGW physically or from the perspective of network topology.
However, in case the UE performs repeated movement over long distances or has stayed in a far-away place after such long-distance movement, the PGW, initially selected, may be far away from a SGW in terms of a distance (a physical or network topological distance). As a result, network efficiency is deteriorated to present problems such as transmission delay of user data or inefficient consumption of network resources within the EPC.
For example, when a passenger to Japan from abroad enters at Narita International Airport, he/she may usually power-up the mobile phone apparatus at the International Airport to attach to EPC. Hence, a PGW located close to Narita International Airport is selected. However, after entrance to Japan, he/she may move to e.g., Tokyo, Osaka, Sapporo or Fukuoka. Hence, after each such movement, the PGW close to the Narita International Airport is no longer the most efficient PGW.